Monthly Archives: November 2013

In theory there is no difference between theory and practice. In practice there is.

ASLlogo300Nothing like starting off with a quote from that great intellectual, Yogi Berra. I promise you – no rants about the over-hyping of Big Data today. No talk of evolution versus revolution. No mention of disruption or innovation or any of that stuff. In fact, the whole point of this blog is the small stuff that we actually do, day in and day out. What I want to talk about is an approach to marketing research that is straightforward, economical, efficient, and likely to lead to that coveted “seat at the table”.

I propose that your job as a researcher on the corporate side is to develop a theory of your brand. Your job as a research supplier is to provide research that helps test that theory. Your job as an academician or developer is to build new measurement instruments that let us test the tenets of these theories better. It’s really that simple.

Here’s what I mean by a theory of your brand:

  • You should know how consumers use your product category, whether it’s a CPG product, a financial service, or a website. Do they use it the way they are supposed to? Are they using it for unintended purposes (think baking soda)? You should likewise know why consumers use your category. Is it a cheaper alternative? More convenient? It works?  Here’s where ethnography and qualitative research are essential.
  • You should know the same things about your brand, how and why people use it. Just as important, you should know why category buyers don’t use your brand, or don’t use it as often as they could. It doesn’t mean you’re going to do anything about it, but you ought to know the reasons why. We have variety in our choices of products so that a product doesn’t have to do everything possible; knowing why shoppers don’t buy your brand may lead you to the next line extension or new brand. Here’s where qual and quant and neuroscience might start to come together.
  • You should know what drives your product’s sales on a day in-day out basis. Are you deal sensitive? Price sensitive? Advertising sensitive? Are you subject to a lot of activity on social media, and if so, does it matter to your brand’s sales? Do you know who your competitors are and whether their activities affect your sales? This last point is not as obvious as it sounds. I’ve had clients tell me that every brand in the category is a competitor (they are not). I’ve had clients think it’s all about one set of brands when it’s really about another set. There are lots of good tools out there to answer this question. This is where Big Data, modeling, and experimentation come in.

The theory is important because, theoretically (sorry about that) your theory of your brand should drive much, if not all, of your future marketing and your future research. Gaps in your theory (and there will be plenty at the start), or places where you’ve filled in with apocrypha or everyone’s best guess, deserve particular attention. This can be a great excuse for rummaging through data you probably already have. For example, we had a client that was consistently disappointed in their new product performances. We found that their forecasts were off because of bad distribution assumptions. They believed, because their sales force believed, that they got 90% ACV distribution in Month 1. We showed them their own syndicated data that suggested an average of 45% ACV in Month 1 – 90% didn’t come until Month 4. When you build a forecast and execute a marketing plan based on bad assumptions, you get bad results – they always came up short of their forecasts. A good theory about how your business works would fix that.

Your theory should also give rise to testable hypotheses, in the form of “if this is true, then doing this should increase sales”. Likewise, your theory can lead you away from research that’s likely to be non-productive or not actionable. Here are two [simplistic, I grant you] examples:

  • You have a brand that you’ve positioned as somewhat upscale in the category – it has benefits that consumers recognize, which is why the keep on buying you. You don’t advertise much – it’s not in your budget. Marketers should be thinking, “How much can we charge for a brand that is upscale”? Very few companies do a lot of pricing research and when they do, it tends to be driven by the necessity to take a price increase due to cost of goods. New product forecasters, on the other hand, tend not to find much of a price impact when you are looking at +/- 5% or so for your average CPG product. Increasing your price might be one heck of a way to increase profits. Worried about a price gap versus competition? Also test the same increases across the other upscale products in the category at the same time.
  • You have a brand that is built on loyal consumers. Shoppers buy you regularly and frequently and your buyers give you a high share of requirements. You know from past research that this is a category that shoppers don’t “shop”; they come into the aisle, buy the products, and then get out. Your marketing team wants to update the package (because marketers always like to update packages before moving on the next brand). Here’s a case where your theory of the brand says a package change is unlikely to help the brand – it could actually hurt it. This is the time to push back and say, “With everything we know about this brand, why would we consider a package change?”

There’s an old quote in Social Psychology by Kurt Lewin, which goes, “There’s nothing so practical as a good theory”.  I’m all for being practical today.

This post originally was published in on 7 November 2013